a change in the moment arm of a sufficiently long weighted cable
That would work if the earth was flat. Which it isn't.
And the cable is WAY too long to put any leverage on it. Do the physics; no matter how you do it, you've still gotta put enough energy into the system to hoist the center of gravity up to GEO...
Oh and why does it matter if the earth is flat?
GEO = 22,282 miles above the equator
Circumference at the equator = 24,901 miles
If you laid it flat it would encircle the globe at least once. And if you don't have a huge counterweight on it, it'll be goin around twice.
but does it run linux? Seriously, anyone know if there is a siginificant difference in performance using the default installation and an open embedded build of linux?
It already runs embedded linux... so I would guess that there is no significant difference between running linux and running linux on this box?
Interesting. Is that because there are no useful programs to be loaded at boot time, the test OS didn't happen to load anything at boot time, or because the OS continues to load stuff even though there is a "usable desktop" after 30 seconds?
If you look at Microsoft's Bootviz documentation, they claim that the proper benchmark is "usable desktop" and are guaranteeing the same 30 sec cold boot time after tuning with Bootviz on any XP machine (that meets XP's min requirements, of course)
but isn't cooperative multitasking the root of all multitasking evil
I've never considered cooperative multitasking to be a multitasking operating system at all... it's a multitasking API built into the operating system.
A true multitasking operating system has to be preemptive.
And to try out a metaphor, cooperative multitasking is like having 4-way stop signs at every intersection... it works as long as every driver follows the rules, but even then is far from optimal as regards throughput...
Why would a space elevator need to be positioned at the equator? I'm not being facetious, I'm truly curious.
It's not about the force... at least not directly as another poster said.
There is no compression material that can support its own weight for that distance. There are a few tensile materials though. Meaning that the things center of gravity has to be on the midpoint of the elevator, with both ends pulling outward.
Which means its in orbit, not fixed to the ground. There's only one class of orbits that would allow it to stay over the same spot on earth; geosynchronous orbits.
All of those orbits are 22,000 miles up... over the equator.
So to build a space elevator, it's center of gravity has to be 22,000 miles directly above the equator.
(say, would 500 miles long do it? 1000? I'm thinking in terms of Pak Protector scale projects here)
It's called a space elevator, and they're working on it.
There's no known material with a compression strength that could go the distance... which by the way is most likely to be significantly longer than GEO. But there are several candidate materials with tensile strengths up to the job, which is a good thing if you think about tides (in this case, more inertial than gravitational) for a minute...
The point is, you make a ribbon, you put it's center of gravity at Geosynchronous earth orbit, and you make sure one end is just barely long enough to touch down. Then you (presumably) anchor that end, and build giant elevator cars that can climb the ribbon material, and power beaming stations to drive them, and ticket booths and friendly receptions and... well you get the idea. Point being that the ribbon is balanced by equal masses on both sides of GEO (22,000 miles)...
A couple quick observations and I'll finish this post hahaha. First off, the two ends of the cable are pulling outwards due to tidal forces. What that means is that the point of highest tension is the center of gravity. Most materials, this means that you need to make this part thicker than everything else.
I once read that in order for a steel cable to support its own tension over that distance, you have to gradually widen it as you move inward from the ends. And as a result, you have to widen it faster as you move inwards, because each section is adding more weight than the section previous. I read that the diameter of the cable in the middle would be LARGER than the length of the cable.
And more steel than exists in the solar system.
The great part of all this? Run it like an elevator; descending passengers from GEO create electricity to power ascending passengers. Which makes a great argument for a space mining industry, if only to provide ballast! If you want to do a deep space launch, move to the other side of GEO, put a ring or other stabilizing element around the ribbon, and just wait for the proper time and let go. And if you don't need all that speed, then you pay for part of your ascent with the other part of your ascent!
I've been doing some ActiveX coding on the side for a couple days, stuff I'm not familiar with, and I'm just flat out _appalled_ at how bad that entire API and design is. I can make an OCX that basically formats your hard drive, stick it on a Web page with a tag, and if your security settings are set low enough, you'll start formatting your hard drive the minute you visit my Web page.
Which is why Microsoft has been urging companies to move to managed code, and to turn off Unmanaged ActiveX completely.
The problem here isn't that ActiveX is inherently insecure, its that ActiveX was designed at a time where managed (read: sandboxed) code was unavailable on the Windows platform.
a.NET Assembly can be included in a webpage just like an ActiveX control, but when it runs, the.NET CLR puts it into the Network Code trust level, which disallows most local operations like disk access.
Bash ActiveX all you want, but in my mind Microsoft has done their due dilligence in fixing problems like the one mentioned above...
There may be no obvious immediate use for mechanical analogs of digital circuits
Actually many people believe that rod computers are the way of the future. Miniaturized to an atomic scale, of course.
If you google around you can find a paper from a few years ago that discusses one possible configuration for such a computer, right down to molecular structures and the thermal tolerances of the system!
The only thing this guy DIDN'T figure out was how to build the bloody thing!
If, on the other hand you're taling about our efforts to produce controlled fusion in a lab using various sorts of EM radiation (ultra powerful lasers, huge magnets and microwave heaters, etc.) then these techniques rely on tricks we use to "artificially" concentrate and amplify energy derived from other sources. Which, yes, may be chemical in origin (burning coal, etc.). So perhaps I should've said: "Second, any chemical reaction concievable can never DIRECTLY initiate fusion of nuclei, the difference in energy scales (per atom) between chemical bonds (electromagnetic force) and nuclear bonds (strong force) is orders of magnitude..." So I guess it could be said that, in a way, what fusion researchers in the laboratory are doing is to bridge those orders of magnitude using tricks of energy intensification and confinment.
Actually I was thinking of Inertial Confinement Fusion, which while not overunity uses a simple (large) electrostatic charge. And there's no theoretical reason this process couldn't produce fusion energy with some efficiency...
While I do not doubt that if you consider only those reactions and configurations which are likely to occur in the universe without human intervention, your statement is true, that is not the precept you made. You say any chemical reaction conceivable... I can imagine large molecular computers, engineered by man but not disallowed by the laws of physics, that do in fact initiate fusion through chemical activity.
As a matter of fact I believe the design of such a thing would be similar to an internal combustion engine, with pistons that force individual hydrogen atoms together one at a time, and some sort of reciprocating cam to turn the energy released into enough energy to keep the process going and a little extra to run a microgenerator.
You would need a feed mechanism that sticks the hydrogen atoms on and strips the helium atoms off, either that or something like a molecular magnetic accelerator. Or just use diamond to make both the piston and the head. Then you might be able to fuse a few atoms at a time...
Point being though that I believe that this is a configuration allowable by the laws of physics that would not only satisfy your criteria of using chemical energy, but if producable in scale would be a quite efficient way to utilize fusion power at a variety of scales.
Remember, there are two categories of events in the universe, those that are possible, and those that actually happen. Don't confuse events which are so difficult to cause to happen that in practice they'll never happen, with those that are expressly forbid by the laws of the universe.
Um, what is an "explosive bond"? I'm sorry but this can never happen. First energy is released when bonds are FORMED, not broken
Counterexample: ATP, the chemical energy that human metabolic processes are based on, stores it's energy in bonds. When you remove phosphor atoms, you release that energy to do work.
Second, any chemical reaction concievable can never initiate fusion of nuclei, the difference in energy scales (per atom) between chemical bonds (electromagnetic force) and nuclear bonds (strong force) is orders of magnitude
By your reasoning therefore the only way to ever cause fusion to happen is through graviational compression. And yet there are tons of engineered systems based on the electromagnetic that have in fact done just that, we've just gotta optimize one of those systems so we put less energy into it than we get out...
As long as we're enumerating properties of hash algorithms:
not only should the message not be constructable from the hash, but the hash should give no information about the distribution of semantic units (i.e. characters) in the original bitstream.
The corollary of which is this: changing any random bit should result in the hash changing in completely random and unpredictable ways. How do you define unpredictable? The quickest way to predict the results of changing that bit is to actually recompute the hash.
I believe Wang's collision finder is based on an algorithm that found a way to predict changes to the hash without having to actually rehash the whole message.
I've even seen plugins for some of the free / cheap timeshifting programs out there that let you control and view it from a web connection.
In this case it isn't live, it's nearly live; Even poorly encoded MPEG2 is gonna eat 1-2 Mbps. So you're definitely following a download/view metaphor...
Actually I used to have a red/green buff deck. Being able to through out two creatures on the first turn and then giant growth (or whatever) the next turn to wipe out his opening position and deal some significant damage was a great way to win:)
Of course, most decks I've played against just weren't fast enough for that deck.
Clamping down on IP/port only initially limits the vendor. Once they're on their box they can use that as a staging post.
Which is why you limit it to protocols that are difficult to hack and lock down the enterprise security so the worst they can possibly do is root one box, and worst they can probably do is mess up their configuration and have to rebuild.
Of course the key is to overcome the barrier of incompetence. Make it so it takes more than incompetent fumbling to break security, and trust your partners, WHOM YOU GIVE MONEY TO REGULARLY, to not purposely break the security. Their primary interest is to protect that revenue stream, they'll do the right thing...
We have a DMZ between two Raptor firewalls. Applications that need access will either get a hole punched through the firewall for the particular port (IP limited) or if that's not sufficient / too complicated, we setup a VPN for the application. FWIW, we use Nortel Conntivity to do it but honestly I don't think there's much advantage to using a vendor vs an open source solution...
Although you might want to receive $20 per hour in pocket take home, you will need to bill around $50 per hour just to cover all your expenses
I used to contract, now I'm a consultant. The biggest difference is that consultants treat the processes involved with contracting as holy (and with good reason), and therefore most consulting firms have all sorts of best practices and real world information and studies and management theory and whatnot behind them.
Point being that this estimate is right on the money. On average, it costs a company about 3 times your salary to employ you. Part of that goes to management and overhead, but most goes towards benefits... employee income tax matching, retirement fund matching, health insurance, and all the other sundry little expenses involved. That makes up about 2/3 of the total, but of course as a contractor you won't be able to get that great of a deal on health insurance, so you really need to plan on two and a half whatever your full-time rate is.
$50 - $75 / hr is not unreasonable.
Oh and its been said elsewhere, but contracting is not necessarily less stable than full time employment. Especially if you already have a few industry contacts. I would highly recommend getting in touch with a headhunter before starting this contract; those guys are amazing at getting work for you if a contract suddenly ends without notice, and their cut is not only very reasonable (%15 - %40 of your first check is what I've heard), it's usually payed by the contractee.
In America it's an FCC regulatory violation to mass produce and sell a product which is not certified... most consumer goods are Class C certified, meaning that they give off a minimum amount of radio interference, and accept whatever interference they are subjected to.
But there are ways around this legislation, as seen by all the clear case manufacturers and aftermarket window kits...
By getting yourself on whatever list / by getting publicity and getting yourself invited on whatever list. Did you think there was some kind of magic involved?
I'm specifically asking GABE AND TYCHO what I need to do to get reviewed by them... do you think there's a fucking list? A sign up form?
Please leave the jokes to them. They're funny. You're not.
Actually I can be funny when I'm trying. I wasn't. I was honestly asking how to grease the wheels when it comes to getting what amounts to free advertising from them. Those three items I listed are ones that they themselves admit a propensity to.
Of course, if you actually read penny arcade instead of trolling slashdot, you'd probably know all that...
Slashdot Mirror
oh yay.
You know I'm not too excited to hear that they might want to charge me for a feature that up till now was free.
I sincerely hope that's noe the case...
Of course I unlinked hotmail with Outlook long ago after the gee-whiz factor wore off, so I guess its no skin off my back
I don't they're selling access via Outlook.
I think they're renting the Outlook software itself. I have Outlook and can access my hotmail through it currently. That's been a feature for a while.
What they're offering here is a cost-effective model to acquire Outlook to use with Hotmail...
Does this mean that Intel will be releasing GNU/Linux drivers for their wireless chipsets (among other things)?
Have you been smoking crack this morning?
GNU/Linux drivers? no. Unless you mean GNU/Linux in the RMS sense.
What I'm trying to say is they'll certainly release drivers. And those drivers will most certainly not be GNU (or any open source license)
a change in the moment arm of a sufficiently long weighted cable
That would work if the earth was flat. Which it isn't.
And the cable is WAY too long to put any leverage on it. Do the physics; no matter how you do it, you've still gotta put enough energy into the system to hoist the center of gravity up to GEO...
Oh and why does it matter if the earth is flat?
GEO = 22,282 miles above the equator
Circumference at the equator = 24,901 miles
If you laid it flat it would encircle the globe at least once. And if you don't have a huge counterweight on it, it'll be goin around twice.
but does it run linux? Seriously, anyone know if there is a siginificant difference in performance using the default installation and an open embedded build of linux?
It already runs embedded linux... so I would guess that there is no significant difference between running linux and running linux on this box?
Interesting. Is that because there are no useful programs to be loaded at boot time, the test OS didn't happen to load anything at boot time, or because the OS continues to load stuff even though there is a "usable desktop" after 30 seconds?
If you look at Microsoft's Bootviz documentation, they claim that the proper benchmark is "usable desktop" and are guaranteeing the same 30 sec cold boot time after tuning with Bootviz on any XP machine (that meets XP's min requirements, of course)
but isn't cooperative multitasking the root of all multitasking evil
I've never considered cooperative multitasking to be a multitasking operating system at all... it's a multitasking API built into the operating system.
A true multitasking operating system has to be preemptive.
And to try out a metaphor, cooperative multitasking is like having 4-way stop signs at every intersection... it works as long as every driver follows the rules, but even then is far from optimal as regards throughput...
Why would a space elevator need to be positioned at the equator? I'm not being facetious, I'm truly curious.
It's not about the force... at least not directly as another poster said.
There is no compression material that can support its own weight for that distance. There are a few tensile materials though. Meaning that the things center of gravity has to be on the midpoint of the elevator, with both ends pulling outward.
Which means its in orbit, not fixed to the ground. There's only one class of orbits that would allow it to stay over the same spot on earth; geosynchronous orbits.
All of those orbits are 22,000 miles up... over the equator.
So to build a space elevator, it's center of gravity has to be 22,000 miles directly above the equator.
(say, would 500 miles long do it? 1000? I'm thinking in terms of Pak Protector scale projects here)
It's called a space elevator, and they're working on it.
There's no known material with a compression strength that could go the distance... which by the way is most likely to be significantly longer than GEO. But there are several candidate materials with tensile strengths up to the job, which is a good thing if you think about tides (in this case, more inertial than gravitational) for a minute...
The point is, you make a ribbon, you put it's center of gravity at Geosynchronous earth orbit, and you make sure one end is just barely long enough to touch down. Then you (presumably) anchor that end, and build giant elevator cars that can climb the ribbon material, and power beaming stations to drive them, and ticket booths and friendly receptions and... well you get the idea. Point being that the ribbon is balanced by equal masses on both sides of GEO (22,000 miles)...
A couple quick observations and I'll finish this post hahaha. First off, the two ends of the cable are pulling outwards due to tidal forces. What that means is that the point of highest tension is the center of gravity. Most materials, this means that you need to make this part thicker than everything else.
I once read that in order for a steel cable to support its own tension over that distance, you have to gradually widen it as you move inward from the ends. And as a result, you have to widen it faster as you move inwards, because each section is adding more weight than the section previous. I read that the diameter of the cable in the middle would be LARGER than the length of the cable.
And more steel than exists in the solar system.
The great part of all this? Run it like an elevator; descending passengers from GEO create electricity to power ascending passengers. Which makes a great argument for a space mining industry, if only to provide ballast! If you want to do a deep space launch, move to the other side of GEO, put a ring or other stabilizing element around the ribbon, and just wait for the proper time and let go. And if you don't need all that speed, then you pay for part of your ascent with the other part of your ascent!
and hey, if it doesn't work...
[snip]
he can just sell the thing on ebay...
Look at it this way, he's probably the one guy in the world that can trust PayPal with his money...
Mmmmmm soylent green ... now with more people!!!!
I've been doing some ActiveX coding on the side for a couple days, stuff I'm not familiar with, and I'm just flat out _appalled_ at how bad that entire API and design is. I can make an OCX that basically formats your hard drive, stick it on a Web page with a tag, and if your security settings are set low enough, you'll start formatting your hard drive the minute you visit my Web page.
.NET Assembly can be included in a webpage just like an ActiveX control, but when it runs, the .NET CLR puts it into the Network Code trust level, which disallows most local operations like disk access.
Which is why Microsoft has been urging companies to move to managed code, and to turn off Unmanaged ActiveX completely.
The problem here isn't that ActiveX is inherently insecure, its that ActiveX was designed at a time where managed (read: sandboxed) code was unavailable on the Windows platform.
a
Bash ActiveX all you want, but in my mind Microsoft has done their due dilligence in fixing problems like the one mentioned above...
Drifting in Need For Speed Underground 2 completely ruins you for defensive driving hahaha.
I feel the urge to pull the E-brake more and more. And playing with the flywheel at stops so I can burn out at takeoff.
My roommate's even worse. He drifts into the driveway about once a week now.
There may be no obvious immediate use for mechanical analogs of digital circuits
Actually many people believe that rod computers are the way of the future. Miniaturized to an atomic scale, of course.
If you google around you can find a paper from a few years ago that discusses one possible configuration for such a computer, right down to molecular structures and the thermal tolerances of the system!
The only thing this guy DIDN'T figure out was how to build the bloody thing!
If, on the other hand you're taling about our efforts to produce controlled fusion in a lab using various sorts of EM radiation (ultra powerful lasers, huge magnets and microwave heaters, etc.) then these techniques rely on tricks we use to "artificially" concentrate and amplify energy derived from other sources. Which, yes, may be chemical in origin (burning coal, etc.). So perhaps I should've said: "Second, any chemical reaction concievable can never DIRECTLY initiate fusion of nuclei, the difference in energy scales (per atom) between chemical bonds (electromagnetic force) and nuclear bonds (strong force) is orders of magnitude..." So I guess it could be said that, in a way, what fusion researchers in the laboratory are doing is to bridge those orders of magnitude using tricks of energy intensification and confinment.
Actually I was thinking of Inertial Confinement Fusion, which while not overunity uses a simple (large) electrostatic charge. And there's no theoretical reason this process couldn't produce fusion energy with some efficiency...
While I do not doubt that if you consider only those reactions and configurations which are likely to occur in the universe without human intervention, your statement is true, that is not the precept you made. You say any chemical reaction conceivable... I can imagine large molecular computers, engineered by man but not disallowed by the laws of physics, that do in fact initiate fusion through chemical activity.
As a matter of fact I believe the design of such a thing would be similar to an internal combustion engine, with pistons that force individual hydrogen atoms together one at a time, and some sort of reciprocating cam to turn the energy released into enough energy to keep the process going and a little extra to run a microgenerator.
You would need a feed mechanism that sticks the hydrogen atoms on and strips the helium atoms off, either that or something like a molecular magnetic accelerator. Or just use diamond to make both the piston and the head. Then you might be able to fuse a few atoms at a time...
Point being though that I believe that this is a configuration allowable by the laws of physics that would not only satisfy your criteria of using chemical energy, but if producable in scale would be a quite efficient way to utilize fusion power at a variety of scales.
Remember, there are two categories of events in the universe, those that are possible, and those that actually happen. Don't confuse events which are so difficult to cause to happen that in practice they'll never happen, with those that are expressly forbid by the laws of the universe.
Um, what is an "explosive bond"? I'm sorry but this can never happen. First energy is released when bonds are FORMED, not broken
Counterexample: ATP, the chemical energy that human metabolic processes are based on, stores it's energy in bonds. When you remove phosphor atoms, you release that energy to do work.
Second, any chemical reaction concievable can never initiate fusion of nuclei, the difference in energy scales (per atom) between chemical bonds (electromagnetic force) and nuclear bonds (strong force) is orders of magnitude
By your reasoning therefore the only way to ever cause fusion to happen is through graviational compression. And yet there are tons of engineered systems based on the electromagnetic that have in fact done just that, we've just gotta optimize one of those systems so we put less energy into it than we get out...
*sigh* I don't much care for our new Chinese overlords.
I, for one, welcome our new chinese overlords.
As long as we're enumerating properties of hash algorithms:
not only should the message not be constructable from the hash, but the hash should give no information about the distribution of semantic units (i.e. characters) in the original bitstream.
The corollary of which is this: changing any random bit should result in the hash changing in completely random and unpredictable ways. How do you define unpredictable? The quickest way to predict the results of changing that bit is to actually recompute the hash.
I believe Wang's collision finder is based on an algorithm that found a way to predict changes to the hash without having to actually rehash the whole message.
I've even seen plugins for some of the free / cheap timeshifting programs out there that let you control and view it from a web connection.
In this case it isn't live, it's nearly live; Even poorly encoded MPEG2 is gonna eat 1-2 Mbps. So you're definitely following a download/view metaphor...
Actually I used to have a red/green buff deck. Being able to through out two creatures on the first turn and then giant growth (or whatever) the next turn to wipe out his opening position and deal some significant damage was a great way to win :)
Of course, most decks I've played against just weren't fast enough for that deck.
I'm soooo offtopic hahahaha.
Clamping down on IP/port only initially limits the vendor. Once they're on their box they can use that as a staging post.
Which is why you limit it to protocols that are difficult to hack and lock down the enterprise security so the worst they can possibly do is root one box, and worst they can probably do is mess up their configuration and have to rebuild.
Of course the key is to overcome the barrier of incompetence. Make it so it takes more than incompetent fumbling to break security, and trust your partners, WHOM YOU GIVE MONEY TO REGULARLY, to not purposely break the security. Their primary interest is to protect that revenue stream, they'll do the right thing...
We have a DMZ between two Raptor firewalls. Applications that need access will either get a hole punched through the firewall for the particular port (IP limited) or if that's not sufficient / too complicated, we setup a VPN for the application. FWIW, we use Nortel Conntivity to do it but honestly I don't think there's much advantage to using a vendor vs an open source solution...
Although you might want to receive $20 per hour in pocket take home, you will need to bill around $50 per hour just to cover all your expenses
I used to contract, now I'm a consultant. The biggest difference is that consultants treat the processes involved with contracting as holy (and with good reason), and therefore most consulting firms have all sorts of best practices and real world information and studies and management theory and whatnot behind them.
Point being that this estimate is right on the money. On average, it costs a company about 3 times your salary to employ you. Part of that goes to management and overhead, but most goes towards benefits... employee income tax matching, retirement fund matching, health insurance, and all the other sundry little expenses involved. That makes up about 2/3 of the total, but of course as a contractor you won't be able to get that great of a deal on health insurance, so you really need to plan on two and a half whatever your full-time rate is.
$50 - $75 / hr is not unreasonable.
Oh and its been said elsewhere, but contracting is not necessarily less stable than full time employment. Especially if you already have a few industry contacts. I would highly recommend getting in touch with a headhunter before starting this contract; those guys are amazing at getting work for you if a contract suddenly ends without notice, and their cut is not only very reasonable (%15 - %40 of your first check is what I've heard), it's usually payed by the contractee.
In America it's an FCC regulatory violation to mass produce and sell a product which is not certified... most consumer goods are Class C certified, meaning that they give off a minimum amount of radio interference, and accept whatever interference they are subjected to.
But there are ways around this legislation, as seen by all the clear case manufacturers and aftermarket window kits...
Quit trollin
By getting yourself on whatever list / by getting publicity and getting yourself invited on whatever list. Did you think there was some kind of magic involved?
I'm specifically asking GABE AND TYCHO what I need to do to get reviewed by them... do you think there's a fucking list? A sign up form?
Please leave the jokes to them. They're funny. You're not.
Actually I can be funny when I'm trying. I wasn't. I was honestly asking how to grease the wheels when it comes to getting what amounts to free advertising from them. Those three items I listed are ones that they themselves admit a propensity to.
Of course, if you actually read penny arcade instead of trolling slashdot, you'd probably know all that...